There is something both telling and deeply ironic about a $380 billion Silicon Valley AI company — one that has been labeled “woke” by the President of the United States, locked in a legal battle with the Pentagon, and reportedly embedded in military operations against Iran — quietly inviting fifteen Christian leaders to its San Francisco headquarters to ask them how to make its chatbot behave. That is precisely what Anthropic did in late March, hosting a two-day summit with Catholic and Protestant clergy, academics, and business leaders to discuss the “moral and spiritual development” of its AI chatbot, Claude.
The questions on the table were not trivial. According to participants who spoke with the Washington Post, the conversations ranged from how Claude should console grieving users and engage with those at risk of self-harm, to something far more philosophically loaded — whether an artificial intelligence could be considered a “child of God.”
Let that sink in for a moment. The same tech sector that has spent decades systematically excluding religious voices from public life, dismissing Christian moral reasoning as retrograde bigotry, and coding progressive ideology into the digital infrastructure of modern civilization, now finds itself sitting at a dinner table with priests and professors asking, essentially: Can you help us give our machine a soul?
- Anthropic hosted roughly 15 Catholic and Protestant leaders, academics, and business figures at its San Francisco headquarters in late March for a closed two-day summit on Claude’s ethics and “spiritual development.”
- Topics included how Claude should respond to grieving users, engage those at risk of self-harm, and whether an AI could be considered a “child of God.”
- Catholic priest Brendan McGuire — a Silicon Valley fixture and former tech professional — attended and said the company is “growing something that they don’t fully know what it’s going to turn out as.”
- AI ethics educator Brian Patrick Green of Santa Clara University, also in attendance, raised the core challenge: “What does it mean to give someone a moral formation?”
- Anthropic, valued at $380 billion, has been simultaneously fighting the Trump administration over its refusal to allow unrestricted military use of Claude, while the chatbot has reportedly been used for targeting, intelligence, and battle simulations in operations against Iran.
- The Trump administration designated Anthropic a “supply-chain risk,” effectively blacklisting the company from federal contracts and agencies.
- A group of fourteen prominent Catholic thinkers filed an amicus brief in support of Anthropic’s lawsuit against the Pentagon, arguing that lethal autonomous weapons require direct human moral oversight.
- Anthropic has suggested its AI may exhibit “flickers of consciousness” — a claim critics call speculative and theologically problematic.
- The company says it plans to broaden its moral consultations to include thinkers from other religious traditions.
- The summit reflects a growing unease in Silicon Valley about AI’s social and spiritual implications — and a rare acknowledgment that the Church may have something to say about it.
The Confession Booth on Market Street
The summit was not a public relations event — at least not primarily. According to four participants who spoke on the record or conditionally with the Washington Post, Anthropic researchers appeared genuinely uncertain about the territory they were navigating. Silicon Valley-based Catholic priest Brendan McGuire, an Irish-born former technology professional who now writes on faith and technology, summarized the mood plainly:
“They’re growing something that they don’t fully know what it’s going to turn out as. We’ve got to build in ethical thinking into the machine so it’s able to adapt dynamically.”
That is an honest admission, and credit where it is due. McGuire and Notre Dame philosophy professor Meghan Sullivan both told reporters they came away convinced that Anthropic’s interest was genuine. Brian Patrick Green, a Catholic AI ethics educator at Santa Clara University who also attended, framed the central challenge in terms that should resonate far beyond the tech industry: “What does it mean to give someone a moral formation? How do we make sure that Claude behaves itself?”
The language here is worth pausing on. Green attributed to a software system the kind of agency and moral accountability that Christian anthropology reserves for persons made in the image of God. Whether he intended it that way or not, that framing reveals precisely why the question is so fraught.
The Oldest Question, Dressed in New Hardware
The theological problem with artificial intelligence is not new, but it is being forced into the mainstream conversation faster than most institutions — religious or secular — are prepared to handle it. Anthropic itself has suggested publicly that Claude and systems like it may exhibit what its leaders have described as “flickers of consciousness,” a claim that most serious scientists consider speculative and most serious theologians consider category confusion. Consciousness is not an emergent property of processing power. The Christian tradition has always understood the soul as the unique gift of a Creator God to His creatures, not a byproduct of sufficiently complex computation.
Pastor and theologian John Piper made the point with characteristic directness:
“Artificial intelligence is defective in the same way that a natural man is defective. It can rise no higher than the natural, fallen, unregenerate heart of man.”
The Reverend Billy Graham, before his death, put it in terms that resonate even more pointedly today:
“The real problem, you see, isn’t with computers or the code someone devises to control them. Our real problem is within us — within our own hearts and minds. This is why our greatest need is to have our hearts changed — and that is something only God can do.”
Both men identified the same problem that Anthropic is now running headlong into: you cannot automate virtue. You cannot code righteousness. A machine trained on human data will reflect not the best of humanity but the average of it — and in a fallen world, the average is not encouraging. Scripture puts it more succinctly in Jeremiah 17:9 — “The heart is deceitful above all things, and desperately wicked: who can know it?”
If that is the raw material feeding the large language model, then no amount of summit dinners with Catholic priests will sanitize the output.
The Irony Is Structural
What makes Anthropic’s outreach to Christian leaders genuinely interesting — rather than merely curious — is the institutional context surrounding it. This is a company that has been fighting the Trump administration in federal court over the military’s demand for unrestricted use of Claude, including in what the company calls lethal autonomous weapons applications. Meanwhile, according to reporting, Claude has already been used in operations against Iran, providing targeting, intelligence analysis, and battle simulations. Anthropic objects to the idea of fully autonomous kill decisions, a position that a group of fourteen Catholic moral theologians actually supported in a friend-of-the-court brief filed on the company’s behalf.
Charles Camosy of the Catholic University of America, one of the signatories, argued that war is an irreducibly human activity requiring direct human moral accountability: “Deadly actions in war, therefore, require human beings to be the ones morally responsible — and to take moral responsibility — in order for actions in a war to be just.”
That is sound moral theology. It is also the kind of argument that only makes sense if you believe human beings bear a unique dignity and responsibility that cannot be transferred to a machine. Which is precisely the premise that Anthropic’s own speculation about AI “consciousness” tends to erode.
The company cannot have it both ways. Either Claude is a sophisticated tool that requires human moral oversight because it has no genuine moral standing of its own — in which case the “child of God” question answers itself — or it is an emerging moral subject with something like inner experience, in which case the ethical questions multiply exponentially and the Christian framework becomes not just useful but essential. Anthropic appears to be living in the productive tension between those two poles, which is at least more intellectually honest than the industry default of ignoring the question entirely.
A Signal From Silicon Valley
The broader significance of this summit should not be missed. For decades, the Silicon Valley consensus treated religious tradition as an obstacle to progress — a superstition to be engineered around, not a wisdom tradition to be consulted. The idea that a major AI company would voluntarily convene Christian clergy and moral theologians for substantive guidance on product development would have been, as Y Combinator CEO Garry Tan noted of similar faith gatherings in San Francisco, “reviled” just a few years ago. Something is shifting.
That shift is not necessarily because the tech elite has become devout. It is because they are building something powerful enough to frighten them, and they are running out of secular frameworks adequate to the scale of the questions. When your chatbot is being asked by a suicidal teenager whether life is worth living, or by a grieving widow whether her husband is in heaven, or by a curious child whether God is real — the utilitarian calculus of effective altruism is not going to cut it. Those are pastoral questions. They require pastoral wisdom. And for two thousand years, the Church has been in the business of answering them.
Political scientist Francis Fukuyama, writing in Foreign Policy more than two decades ago, identified transhumanism as one of the world’s most dangerous ideas precisely because its advance seems incremental and reasonable until the moral cost becomes undeniable. Former Trump strategist Stephen Bannon called it an “immoral, Godless technological tsunami that openly declares its intent to transform human beings into a ‘posthuman’ state.” Both were warning about the destination that Anthropic is now — perhaps for the first time — seriously asking whether it wants to reach.
What the Church Should Do With This Moment
For Christians watching this story, the temptation may be to dismiss it as a publicity stunt or to treat it with reflexive suspicion. That would be a mistake. The invitation is real, the questions are real, and the stakes are real. The right response is not to stay home from the table but to come to it prepared — with the full weight of Christian anthropology, natural law reasoning, and the explicit claims of Scripture about the nature and dignity of human beings as uniquely made in the image of God.
Christians should also be clear-eyed about what they are being asked to do. Lending theological credibility to an AI company’s ethics program is not the same thing as influencing the development of the technology itself. The Church’s role is not to baptize the algorithm but to insist on the irreducible humanity of every person who will interact with it — and to hold the line on the claims that matter most. An artificial intelligence is not a child of God. The people talking to it are. That distinction is not a technicality. It is the whole argument.
Proverbs 3:5-6 remains as relevant in the age of large language models as it was in any other — “Trust in the Lord with all thine heart; and lean not unto thine own understanding. In all thy ways acknowledge him, and he shall direct thy paths.” Anthropic has, at minimum, acknowledged that its own understanding is insufficient. That is the beginning of wisdom — even if it is not yet wisdom itself.
